### Install Khoj Assistant (Quick Start)

Source: https://github.com/khoj-ai/khoj/wiki/Install-Khoj-on-Windows-🪟

Install the Khoj assistant package using pip. This is the primary command for the quick start installation.

```bash
pip install khoj-assistant
```

--------------------------------

### Start Khoj (Quick Start)

Source: https://github.com/khoj-ai/khoj/wiki/Install-Khoj-on-Windows-🪟

Launch the Khoj application after installation. This command assumes Khoj has been successfully installed.

```bash
khoj
```

--------------------------------

### Install Project Dependencies

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/desktop/README.md

Installs all development dependencies for the project. Run this before starting the application.

```bash
yarn install
```

--------------------------------

### Setup Development Environment and Git Hooks

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Install Khoj, set up Git hooks for automatic code formatting and type checking on commit and push. Use the `--full` flag to also install Khoj Desktop and Obsidian client apps.

```shell
./scripts/dev_setup.sh
```

--------------------------------

### Minimal Install Khoj.el with use-package

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/clients/emacs.md

Configure Khoj.el using the `use-package` macro for minimal setup, including binding a key for quick access. Assumes MELPA Stable.

```elisp
;; Install Khoj client from MELPA Stable
(use-package khoj
  :ensure t
  :pin melpa-stable
  :bind ("C-c s" . 'khoj)
  :config (setq khoj-api-key "YOUR_KHOJ_CLOUD_API_KEY"
                khoj-server-url "https://app.khoj.dev"))
```

--------------------------------

### Setup Khoj with Docker

Source: https://context7.com/khoj-ai/khoj/llms.txt

Sets up Khoj using Docker for multi-user, production-ready deployment. Requires creating a configuration directory, downloading docker-compose.yml, and editing it with secure credentials and API keys. Then, start the services with 'docker-compose up'.

```bash
mkdir ~/.khoj && cd ~/.khoj
wget https://raw.githubusercontent.com/khoj-ai/khoj/master/docker-compose.yml

# Edit docker-compose.yml:
#   KHOJ_ADMIN_PASSWORD=<secure>
#   KHOJ_DJANGO_SECRET_KEY=<secure>
#   OPENAI_API_KEY=<your-key>          # or ANTHROPIC_API_KEY / GEMINI_API_KEY
#   # Uncomment OPENAI_BASE_URL to use Ollama on the host

docker-compose up
```

--------------------------------

### Run the Application Locally

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/desktop/README.md

Starts the Khoj AI desktop application. Ensure all dependencies are installed first.

```bash
yarn start
```

--------------------------------

### Install WSL2 and Docker on Windows

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install WSL2 and Docker Desktop with WSL2 backend on Windows. Restart your machine after WSL2 installation.

```shell
# Run in PowerShell
wsl --install
```

--------------------------------

### Start Local Development Server

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/README.md

Starts a local development server for live preview. Changes are reflected without server restart.

```bash
$ yarn start
```

--------------------------------

### Direct Install Khoj.el

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/clients/emacs.md

Install the Khoj Emacs client directly and set your API key and server URL. This is the most straightforward installation method.

```elisp
;; Install Khoj.el
M-x package-install khoj

; Set your Khoj API key
(setq khoj-api-key "YOUR_KHOJ_CLOUD_API_KEY")
(setq khoj-server-url "https://app.khoj.dev")
```

--------------------------------

### Install Project Dependencies

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/web/README.md

Installs all necessary dependencies for the project using Bun. Run this command first to set up your development environment.

```bash
bun install
```

--------------------------------

### Install LiteLLM

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/advanced/litellm.md

Install LiteLLM with the necessary proxy dependencies. This command is used to set up the LiteLLM proxy server.

```bash
pip install litellm[proxy]
```

--------------------------------

### Install Khoj using Quelpa in Emacs

Source: https://github.com/khoj-ai/khoj/blob/master/tests/data/org/interface_emacs_readme.org

Install the khoj package using Quelpa, specifying the URL for the khoj.el file. Ensure Quelpa and quelpa-use-package are installed.

```elisp
(use-package khoj
  :quelpa (khoj :fetcher url :url "https://raw.githubusercontent.com/khoj-ai/khoj/master/interface/emacs/khoj.el")
  :bind ("C-c s" . 'khoj))
```

--------------------------------

### Install ToDesktop CLI

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/desktop/README.md

Installs the ToDesktop command-line interface globally. This is a prerequisite for deploying the Electron app.

```bash
yarn global add @todesktop/cli
```

--------------------------------

### Install Dependencies and Build Plugin

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/obsidian/README.md

Run these commands in the `src/interface/obsidian` directory to install project dependencies and build the Obsidian plugin.

```bash
yarn install
yarn build
```

--------------------------------

### Run Docker Compose

Source: https://github.com/khoj-ai/khoj/blob/master/src/khoj/app/README.md

Use this command to start the Khoj application using Docker Compose. Ensure you have Docker and Docker Compose installed.

```bash
docker-compose up
```

--------------------------------

### Start Khoj Docker with Ollama

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/advanced/ollama.mdx

Run this command in the directory containing your Khoj docker-compose.yml file after uncommenting OPENAI_BASE_URL. This starts Khoj and integrates it with your host's Ollama.

```bash
# run below command in the directory where you downloaded the Khoj docker-compose.yml
docker-compose up
```

--------------------------------

### Install Khoj Server (Pip - Windows CPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Windows using CPU.

```shell
# Install Khoj
py -m pip install 'khoj[local]'
```

--------------------------------

### Install Dependencies

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/README.md

Installs project dependencies using Yarn.

```bash
$ yarn
```

--------------------------------

### Start Khoj with Docker Compose

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Start the Khoj server using Docker Compose. Ensure you are in the same directory as your docker-compose.yml file.

```shell
cd ~/.khoj
docker-compose up
```

--------------------------------

### Install Docker and Docker Compose on MacOS

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Use Homebrew to install Docker and Docker Compose on MacOS. Ensure Docker Desktop is also installed.

```shell
brew install --cask docker
brew install docker-compose
```

--------------------------------

### Install Khoj Server (Pip - Windows Vulkan GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Windows, enabling VULKAN GPU acceleration. Set the CMAKE_ARGS environment variable before installation.

```shell
# 1. To use VULCAN GPU
$env:CMAKE_ARGS = "-DGGML_VULKAN=on"
# 2. Install Khoj
py -m pip install 'khoj[local]'
```

--------------------------------

### Install Khoj Project

Source: https://github.com/khoj-ai/khoj/blob/master/tests/data/org/main_readme.org

Clone the repository, create a Conda environment, and activate it to set up the project.

```shell
git clone https://github.com/khoj-ai/khoj && cd khoj
conda env create -f environment.yml
conda activate khoj
```

--------------------------------

### Install Khoj Server (Pip - Linux Vulkan GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Linux, enabling VULKAN GPU acceleration. Set CMAKE_ARGS and FORCE_CMAKE before installation.

```shell
CMAKE_ARGS="-DGGML_VULKAN=on" FORCE_CMAKE=1 python -m pip install 'khoj[local]'
```

--------------------------------

### Start LiteLLM Proxy with Mistral Tiny

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/advanced/litellm.md

Start the LiteLLM proxy server, specifying the model to use and optionally dropping parameters. Ensure your Mistral API key is exported as an environment variable.

```bash
export MISTRAL_API_KEY=<MISTRAL_API_KEY>
litellm --model mistral/mistral-tiny --drop_params
```

--------------------------------

### Install Rust for Tokenizer Dependency

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Use these commands to install Rust if `pip install khoj` fails due to the `tokenizers` dependency requiring Rust compilation. This is a common fix for build errors.

```shell
brew install rustup
rustup-init
source ~/.cargo/env
```

--------------------------------

### Install Khoj and Dependencies (Windows)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Clones the Khoj repository, sets up a Python virtual environment, and installs all necessary dependencies including extras for development on Windows.

```shell
# Get Khoj Code
git clone https://github.com/khoj-ai/khoj && cd khoj

# Create, Activate Virtual Environment
python3 -m venv .venv && .venv\Scripts\activate

# Install Khoj for Development
uv sync --all-extras

```

--------------------------------

### Install Khoj and Dependencies (Linux)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Clones the Khoj repository, sets up a Python virtual environment, and installs all necessary dependencies including extras for development on Linux.

```shell
# Get Khoj Code
git clone https://github.com/khoj-ai/khoj && cd khoj

# Create, Activate Virtual Environment
python3 -m venv .venv && source .venv/bin/activate

# Install Khoj for Development
uv sync --all-extras

```

--------------------------------

### Install Khoj Server (Pip - Windows NVIDIA GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Windows, enabling NVIDIA (CUDA) GPU acceleration. Set the CMAKE_ARGS environment variable before installation.

```shell
# 1. To use NVIDIA (CUDA) GPU
$env:CMAKE_ARGS = "-DGGML_CUDA=on"
# 2. Install Khoj
py -m pip install 'khoj[local]'
```

--------------------------------

### Install Khoj with Production Dependencies

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/advanced/authentication.mdx

Use this command to install the production version of the Khoj package, which is required for Google OAuth integration.

```bash
pip install khoj[prod]
```

--------------------------------

### Start Khoj Application

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Run the Khoj application with verbose logging enabled.

```bash
khoj -vv
```

--------------------------------

### Install Khoj Package

Source: https://github.com/khoj-ai/khoj/blob/master/tests/data/markdown/main_readme.md

Install the Khoj package using pip. Ensure Python 3 and Miniconda are set up as dependencies.

```bash
pip install khoj
```

--------------------------------

### Install Khoj with Pip (Linux CPU)

Source: https://context7.com/khoj-ai/khoj/llms.txt

Installs Khoj for Linux systems with CPU-only support using pip. Assumes local, single-user, anonymous access.

```bash
pip install 'khoj[local]'
```

--------------------------------

### Start Khoj Server

Source: https://github.com/khoj-ai/khoj/blob/master/src/khoj/app/README.md

Starts the Khoj application server using FastAPI with Django as the ORM backend. The '--anonymous-mode' flag bypasses authentication for local development.

```bash
python3 src/khoj/main.py --anonymous-mode
```

--------------------------------

### Install Khoj and Dependencies (MacOS)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Clones the Khoj repository, sets up a Python virtual environment, and installs all necessary dependencies including extras for development on MacOS.

```shell
# Get Khoj Code
git clone https://github.com/khoj-ai/khoj && cd khoj

# Create, Activate Virtual Environment
python3 -m venv .venv && source .venv/bin/activate

# For MacOS or zsh users run this
uv sync --all-extras

```

--------------------------------

### Install Khoj with Pip (macOS Apple Silicon)

Source: https://context7.com/khoj-ai/khoj/llms.txt

Installs Khoj with Metal support for macOS Apple Silicon using pip. Assumes local, single-user, anonymous access.

```bash
CMAKE_ARGS="-DGGML_METAL=on" pip install 'khoj[local]'
```

--------------------------------

### Run Docker Compose

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Start the Khoj server and database using Docker Compose in detached mode.

```shell
docker-compose up -d
```

--------------------------------

### Start Khoj Server

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Start the Khoj server with an embedded database and anonymous mode enabled. This command is used after installing Khoj via pip.

```shell
USE_EMBEDDED_DB="true" khoj --anonymous-mode
```

--------------------------------

### Set Ollama Environment Variable for Pip Install

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/advanced/ollama.mdx

Before starting Khoj with pip, set the OPENAI_BASE_URL environment variable to point to your Ollama instance. Then, start Khoj in anonymous mode.

```bash
export OPENAI_BASE_URL="http://localhost:11434/v1/"
khoj --anonymous-mode
```

--------------------------------

### Standard Install Khoj.el with custom indexing

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/clients/emacs.md

Set up Khoj.el with `use-package`, specifying custom directories and files to be indexed. This allows for more targeted indexing of your notes.

```elisp
;; Install Khoj client from MELPA Stable
(use-package khoj
  :ensure t
  :pin melpa-stable
  :bind ("C-c s" . 'khoj)
  :config (setq khoj-api-key "YOUR_KHOJ_CLOUD_API_KEY"
                khoj-server-url "https://app.khoj.dev"
                khoj-index-directories '("~/docs/org-roam" "~/docs/notes")
                khoj-index-files '("~/docs/todo.org" "~/docs/work.org")))
```

--------------------------------

### Build and Serve Production Files

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/web/README.md

Builds the project for production and serves the static files. This is useful for testing the production build locally before deployment.

```bash
bun export
```

--------------------------------

### Install Khoj with Pipx

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install Khoj using pipx to avoid dependency conflicts with other Python packages. This is recommended when encountering issues during pip installation.

```shell
pipx install khoj
```

--------------------------------

### Build and Serve Production Files (Windows)

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/web/README.md

Builds the project for production and serves the static files. This command is specifically for Windows users.

```bash
bun windowsexport
```

--------------------------------

### Build the Electron App

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/desktop/README.md

Triggers builds for all platforms after prompting for login. Ensure the ToDesktop CLI is installed and configured.

```bash
todesktop build
```

--------------------------------

### Get Notion Configuration

Source: https://context7.com/khoj-ai/khoj/llms.txt

Get the current Notion integration configuration.

```APIDOC
## GET /api/content/notion

### Description
Get current Notion configuration.

### Method
GET

### Endpoint
/api/content/notion

### Request Example
```bash
curl -s "http://localhost:42110/api/content/notion" \
  -H "Authorization: Bearer <API_KEY>"
```
```

--------------------------------

### Build Front-end Assets

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Navigates to the web interface directory, installs front-end dependencies using Bun, and exports the production build of the front-end assets.

```shell
cd src/interface/web/
bun install
bun export

```

--------------------------------

### Upgrade Khoj Installation

Source: https://context7.com/khoj-ai/khoj/llms.txt

Provides commands to upgrade Khoj installations, both for pip and Docker.

```bash
pip install --upgrade khoj          # pip
```

```bash
docker-compose up --build           # Docker
```

--------------------------------

### Install Khoj Server (Pip - Linux AMD GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Linux, enabling AMD (ROCm) GPU acceleration. Set CMAKE_ARGS and FORCE_CMAKE before installation.

```shell
CMAKE_ARGS="-DGGML_HIPBLAS=on" FORCE_CMAKE=1 python -m pip install 'khoj[local]'
```

--------------------------------

### Install Khoj Server (Pip - Linux NVIDIA GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Linux, enabling NVIDIA (CUDA) GPU acceleration. Set CMAKE_ARGS and FORCE_CMAKE before installation.

```shell
CMAKE_ARGS="-DGGML_CUDA=on" FORCE_CMAKE=1 python -m pip install 'khoj[local]'
```

--------------------------------

### Install Khoj Server (Pip - Windows AMD GPU)

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/get-started/setup.mdx

Install the Khoj server package with local support on Windows, enabling AMD (ROCm) GPU acceleration. Set the CMAKE_ARGS environment variable before installation.

```shell
# 1. To use AMD (ROCm) GPU
$env:CMAKE_ARGS = "-DGGML_HIPBLAS=on"
# 2. Install Khoj
py -m pip install 'khoj[local]'
```

--------------------------------

### Run Development Server

Source: https://github.com/khoj-ai/khoj/blob/master/src/interface/web/README.md

Starts the Next.js development server. This command allows for hot-reloading and live updates as you edit your code.

```bash
bun dev
```

--------------------------------

### Install Postgres with PgVector on Debian/Ubuntu

Source: https://github.com/khoj-ai/khoj/blob/master/src/khoj/app/README.md

Installs PostgreSQL and the PgVector extension on Debian or Ubuntu systems. This is a prerequisite for the Khoj database.

```bash
sudo apt install -y postgresql-common
sudo /usr/share/postgresql-common/pgdg/apt.postgresql.org.sh
sudo apt install postgres-16 postgresql-16-pgvector
```

--------------------------------

### Run Development Server for Obsidian Plugin

Source: https://github.com/khoj-ai/khoj/blob/master/documentation/docs/contributing/development.mdx

Execute this command in the `src/interface/obsidian` directory to start the development server. It will continuously rebuild the plugin as code changes are made.

```shell
yarn dev
```